Garrett, R., et al. (2013). Discrimination of arabic coffee cultivars by electrospray ionization Fourier Tansform ion cyclotron resonance mass spectrometry and chemometrics. Food Sci Technol-LEB. 50, 496-502. DOI Link
Introduction
- Arabica coffee is the most popular and high quality coffee variety produced.
- Analysis of coffer is difficult because both genetic and environmental factors affect many aspects of the final product.
- The excellent mass resolution and mass accuracy of FT-ICR MS makes it a powerful tool in studying metabolites in order to characterize and differentiate samples.
- This group studied beans of several genetic profiles, grown in two Brazilian regions grown under similar soil
and climate conditions.
Material and methods
- A description of the sources, genetic backgrounds, growing conditions, harvesting procedures, and storage methods of the samples is provided.
- Sample preparation to prepare the beans for the mass spectrometer is provided.
- Instrumental details for a FT-ICR MS with an ESI source are provided.
- Data analysis procedures are also provided.
- Details about the multivariate analysis are provided, including the generation of the matrices for PCA and
PLS-DA.
Results and discussion
- Generally, the same m/z peaks were seen in all the samples. However, the intensities varied between the various
sources in a statistically significant and reliable manner.
- 20 compounds were identified using a coffee library by matching the mass to within 2 ppm and matching the
isotope patterns with calculated patterns. All identified compounds have been previously identified in coffee samples
in the literature.
- A discussion of fragmentation of several important molecules in the analysis is provided.
- Two key compounds involved in the multivariate analysis were not able to be identified within previously
discussed accuracy of limits.
- PCA was used. PCA find the largest sources of variation within a data set, then assigns a score for each
source.
- PLS-DA introduces a "dummy matrix" to obtain a numerical measure of the differences between the various data
sets of beans with different genetic background and growing regions.
- Details of important (high variation) peaks are provided in the next three paragraphs. The compounds are
identified and the degree of differences are listed. Additionally, details of the PCA and PCL-DA processes are
included.
- The genetic profiles were able to be differentiated with these methods.
- The growing region was not discriminated in all cases. Some were separated, but others were not.
- Depending on the region, the intensities of [M+H]- peaks varied and was used to identify between the genetic
backgrounds.
- The climate probably effected the chemical levels in the beans.
- The key chemicals for discrimination are listed in this paragraph.
Conclusion
- Sample preparation, data collection and data analysis methods are summarized.
- An application of this data is identifying where beans were grown for certification procedures in the future.
Garrett, R., et al. (2013). Discrimination of arabic coffee cultivars by electrospray ionization Fourier Tansform ion cyclotron resonance mass spectrometry and chemometrics. Food Sci Technol-LEB. 50, 496-502. DOI Link
Introduction
- Arabica coffee is the most popular and high quality coffee variety produced.
- Analysis of coffer is difficult because both genetic and environmental factors affect many aspects of the final product.
- The excellent mass resolution and mass accuracy of FT-ICR MS makes it a powerful tool in studying metabolites in order to characterize and differentiate samples.
- This group studied beans of several genetic profiles, grown in two Brazilian regions grown under similar soil
and climate conditions.
Material and methods
- A description of the sources, genetic backgrounds, growing conditions, harvesting procedures, and storage methods of the samples is provided.
- Sample preparation to prepare the beans for the mass spectrometer is provided.
- Instrumental details for a FT-ICR MS with an ESI source are provided.
- Data analysis procedures are also provided.
- Details about the multivariate analysis are provided, including the generation of the matrices for PCA and
PLS-DA.
Results and discussion
- Generally, the same m/z peaks were seen in all the samples. However, the intensities varied between the various
sources in a statistically significant and reliable manner.
- 20 compounds were identified using a coffee library by matching the mass to within 2 ppm and matching the
isotope patterns with calculated patterns. All identified compounds have been previously identified in coffee samples
in the literature.
- A discussion of fragmentation of several important molecules in the analysis is provided.
- Two key compounds involved in the multivariate analysis were not able to be identified within previously
discussed accuracy of limits.
- PCA was used. PCA find the largest sources of variation within a data set, then assigns a score for each
source.
- PLS-DA introduces a "dummy matrix" to obtain a numerical measure of the differences between the various data
sets of beans with different genetic background and growing regions.
- Details of important (high variation) peaks are provided in the next three paragraphs. The compounds are
identified and the degree of differences are listed. Additionally, details of the PCA and PCL-DA processes are
included.
- The genetic profiles were able to be differentiated with these methods.
- The growing region was not discriminated in all cases. Some were separated, but others were not.
- Depending on the region, the intensities of [M+H]- peaks varied and was used to identify between the genetic
backgrounds.
- The climate probably effected the chemical levels in the beans.
- The key chemicals for discrimination are listed in this paragraph.
Conclusion
- Sample preparation, data collection and data analysis methods are summarized.
- An application of this data is identifying where beans were grown for certification procedures in the future.